Hacker News Reader

Comment by mikeaskew4 | Hacker News Reader

Thumbs down. Great design is original thought. AI is wholly incapable of that.

Go ahead and roast me.

> Great design is original thought

I don't agree. For novel use cases, yes there's some truth to that. But consistency is huge in a UX. If basic controls work well for a situation, they should be used. Designers should not be getting "creative" or "original" for those sorts of things.

I don't think you are really disagreeing. I'd guess mikeaskew4 would include the constraints of following platform conventions / HIGs and using native widgets for any GUI. Those constraints do not preclude original thought.

Transparently, I'm not a designer, I'm a biologist. That said, the things I want designed by Claude don't need great design. I need a slide format that is consistent from one study to the next so the reader can follow. I need a tool that tracks the number of mice in each lab and flags if someone is using more resources than we expect. I need a personal site that is easy to work with that tracks my pet geckos' feed and environment.

If I have a product out of my lab that makes it to human trials, there will be a full team of marketers and designers tasked to the brand image.

I doubt you’d be singing the same tune if Anthropic released Claude Researcher and a bunch of people were saying “I’m a designer, not a biologist. I don’t need great research. I just need X”

Ironically I think AI will replace researchers before it replaces artists.

I'm not the one saying my job is uniquely human. Last week, I watched a video of a vendor's new facility that entirely automates antibody production and screening lab. The machines run literally every step from picking antibodies, sequencing, scaling up, and in vitro screening without a human interfering or donning a labcoat. A machine that is essentially a nicer roomba fetches the flask and drives it to where the shakers are and puts it on an empty shaker spot. I have no doubt the lab tasks I do can be automated. Fortunately for me, a large portion of my job at the moment is in handling animals, and there are much higher barriers there, but again, I don't think that part of my job makes me irreplaceable.

> Great design is original thought. AI is wholly incapable of that

This is entirely accurate, however I fear there's a lack of perspective:

If you're in the middle of the desert and need to sit down, that random rock looks and feels great because there's nothing even close, around!

One issue that a lot of experts fail to recognize is that "great" is relative: It's not apparent to the experts because they are only pulled in when their expertise is needed. Most of the time when experts are pulled in, requirements are clear, you have traction, scale and now you need to optimize.

Once you're spoiled for choices, you have lots of options and then that random rock doesn't look appealing at all: now you're considering other factors like budget - IKEA vs Adirondack.

What AI is making a huge difference are places where "great" isnt that valuable:

- people in the desert: Someone wants to track what words their toddler is saying or their groceries or how much kitty litter they should buy soon and Claude will spit something out reasonable even if it makes the skin of experts crawl.

- commodity and bean counters: in cut throat industries like power or insurance, it's all commodity services competing on price. Most people arent going to pay a premium for a better looking, more intuitive insurance app. It just needs to not suck and fall over. Or you're making a knockoff of an existing, well understood product

As if original thought occurs out of thin air. AI is definitely capable to take aspects of design and apply it elsewhere.

Maybe you haven't noticed, but most of the internet doesn't seem to care about great design.

Also, a lot of very good software developers are bad at design and unwilling or unable to pay for a designer. This will be an improvement for them.

That’s pretty true. Just look at this sites design

Having furniture made by a master designer and artisan woodworker is great. But sometimes, you don't have the budget for that and Ikea furnitures does the job just fine. Most of the time, actually.

IKEA hires many great designers for its furniture though.

You're absolutely correct.

But the mass market (who this is ultimately for) doesn't care about great design. They care about "seeing something on the screen." If they can get something that looks 80-90% aligned with what they observe to be modern design, they won't think twice (even if the end result is clunky or not on par with what a professional designer would produce). It's the Ikea Effect on steroids.

Tell this to the 99% of designers who are designing the 5th page in some RBAC modal or some obscure settings page. Design is like code - there are a few people doing really groundbreaking stuff, but vastly more doing the utilitarian plumbing everyone depends on.

My default position: If an LLM can create it, we probably don't actually need it.

If AI can create it- it's probably already been created and thought of before?

You can let the LLM create slop for you, sure. But only amateurs are using it for that. You’ll be much happier if you treat it as a tool and use it like any other, a force multiplier to take your ideas and creations and pushes them further along faster.

If you treat it like a black box used to outsource your own thinking, you are holding it wrong.

It’s honestly been my experience. I think I’ve vibe coded one thing that I’ve found to be genuinely useful to me, everything else has ended up as useless slop given time.

I think we probably need less software, but higher quality, not more. Unfortunately AI only goes in one direction…

This is just going to chop the bottom end off, same as with software. If you are great you get to keep your job.

The catch is that the person making the decision might not know or care about the difference.

Have you seen many original designs in the past 15 years?

Most applications just need good enough design.

Sure, great design is original thought.

Is there also a place in the world for not-great-but-good-enough design?

> Great design is original thought

not really, great design in a web application is no surprises.

Great designers will make great designs no matter what tools you give them.

Human inputs their idea. AI helps bring it to light faster. Human iterates.

Still human?

i dont disagree. but there is a difference between great design and entirely acceptable and shippable design.

Reply on news.ycombinator.com